© Jitesh Khurkhuriya – Azure ML Online Course

What is Microsoft Azure ML?

© Jitesh Khurkhuriya – Azure ML Online Course

 Azure Machine Learning
• Cloud based predictive analytics service

• Provides tools to create complete machine learning solution in the cloud

• Quick model creation and deployment using Azure ML Studio

• Allows Models to be deployed as web services

• Provides a large library of Pre-Built Machine learning algorithms and Modules

• Allows for extending your models with custom built R and Python code

© Jitesh Khurkhuriya – Azure ML Online Course

 Azure ML Studio Overview
Access Azure anytime, Projects, Experiments, Models, Datasets
anywhere using internet

• Data processing
• Feature Engineering
• Model Selection
• Model Scoring and Training
• Test and deploy models

Deploy and Consume using

Web Services
• Via HTTP
Export and Import
• Hive
data to Azure Cloud
• Azure SQL DB
or ML Studio
• Data Feed
• On Premise Data
• Azure Document DB
© Jitesh Khurkhuriya
• – Azure
Azure ML OnlineStorage
BLOB Course
Create Your Azure ML Account

Disclaimer: Some Screens may have changed as Microsoft keeps updating the portal and user experience.

© Jitesh Khurkhuriya – Azure ML Online Course

www.microsoft.com

© Jitesh Khurkhuriya – Azure ML Online Course

© Jitesh Khurkhuriya – Azure ML Online Course
© Jitesh Khurkhuriya – Azure ML Online Course
© Jitesh Khurkhuriya – Azure ML Online Course
© Jitesh Khurkhuriya – Azure ML Online Course
© Jitesh Khurkhuriya – Azure ML Online Course
© Jitesh Khurkhuriya – Azure ML Online Course
© Jitesh Khurkhuriya – Azure ML Online Course
© Jitesh Khurkhuriya – Azure ML Online Course
Welcome to Machine Learning

© Jitesh Khurkhuriya – Azure ML Online Course

studio.azureml.net

© Jitesh Khurkhuriya – Azure ML Online Course

Welcome to Azure ML Studio

© Jitesh Khurkhuriya – Azure ML Online Course

Workflow of Azure ML
project/experiment

© Jitesh Khurkhuriya – Azure ML Online Course

 Workflow of Azure ML

GET THE DATA

PREPARE THE DATA

FEATURE SELECTION

CHOOSE AND APPLY LEARNING ALGORITHMS

TRAIN AND EVALUATE THE MODEL

© Jitesh Khurkhuriya – Azure ML Online Course

 Workflow of Azure ML

GET THE DATA

PREPARE THE DATA

Feature Selection

CHOOSE AND APPLY LEARNING

ALGORITHMS

TRAIN AND EVALUATE THE MODEL

© Jitesh Khurkhuriya – Azure ML Online Course

 Workflow of Azure ML

GET THE DATA

PREPARE THE DATA

Feature Selection

CHOOSE AND APPLY LEARNING

ALGORITHMS

TRAIN AND EVALUATE THE MODEL

© Jitesh Khurkhuriya – Azure ML Online Course

 Workflow of Azure ML

GET THE DATA

PREPARE THE DATA

Feature Selection

CHOOSE AND APPLY LEARNING

ALGORITHMS

TRAIN AND EVALUATE THE MODEL

© Jitesh Khurkhuriya – Azure ML Online Course

 Workflow of Azure ML

GET THE DATA

PREPARE THE DATA

Feature Selection

CHOOSE AND APPLY LEARNING

ALGORITHMS

TRAIN AND EVALUATE THE MODEL

© Jitesh Khurkhuriya – Azure ML Online Course

 Workflow of Azure ML

GET THE DATA

PREPARE THE DATA

Feature Selection

CHOOSE AND APPLY LEARNING

ALGORITHMS

TRAIN AND EVALUATE THE MODEL

© Jitesh Khurkhuriya – Azure ML Online Course

Azure ML Algorithm CheatSheet

© Jitesh Khurkhuriya – Azure ML Online Course

 Predicting Categories MULTI-CLASS CLASSIFICATION

Outcome is Binary? Yes or No,

Pass or Fail, Success or failure?
Start

TWO-CLASS CLASSIFICATION

Accuracy, Fast Two-Class Decision

>100 Features, Forest
Two Class SVM Training
Linear Model
Accuracy, Fast Two-Class Boosted
Two-Class Averaged Fast Training, Training, LargeM Decision Tree
Perceptron Linear Model
Accuracy, SmallM Two Class Decision
Two Class Logistic Fast Training, Jungle
Regression Linear model
>100 Features Two Class Locally Deep
Two Class Bayes Fast Training, SVM
Point Machine Linear Model
Accuracy, Long Two Class Neural
Training Times Network
© Jitesh Khurkhuriya – Azure ML Online Course
Predicting Categories MULTI-CLASS CLASSIFICATION

Fast Training, Linear Model Multi-Class Logistic Regression

Accuracy, Long Training Times Multi-Class Neural Network

Accuracy, Fast Training Multi-Class Decision Forest

Outcome has multiple Accuracy, Small Memory Footprint Multi-Class Decision Jungle
possibilities? Customer
categories etc? Start Depends on Two-Class One-V-All Multiclass

TWO-CLASS CLASSIFICATION

© Jitesh Khurkhuriya – Azure ML Online Course

REGRESSION Predicting Continuous Value
Data in Rank Order
Start
Ordinal Regression
categories

Poisson Regression Predicting Event Counts

Predicting a Such as Sales or revenue

Fast Forest Quantile Regression Distribution forecast, stock price, Loyalty
Fast Training, Linear score etc
Linear Regression
Model

Linear Model, Small

Bayesian Linear Regression
datasets

Neural Network Regression Accuracy, Long Training

Time

Decision Forest Regression Accuracy, Fast Training

Accuracy, Fast Training,

Boosted Decision Tree Regression
large Memory
© Jitesh Khurkhuriya – Azure ML Online Course
ANOMALY DETECTION

One Class SVM > 100 Features

PCA Based Anomaly Detection Fast Training

Finding Unusual Data Points

Start

Such as Fraudulent
Transactions, abnormal reading
from machines etc

© Jitesh Khurkhuriya – Azure ML Online Course

 What to consider while choosing an algorithm?

Predicting Categories

CLUSTERING
Predicting Continuous Value
K-MEANS

Finding Unusual Data Points

Discovering Structure Start

© Jitesh Khurkhuriya – Azure ML Online Course

 What to consider while choosing an algorithm?

Accuracy
Predicting Categories
Training Time
Predicting Values
Number of Features
Finding Unusual Data Points
Number of Parameters
Discovering Structure
Linearity

© Jitesh Khurkhuriya – Azure ML Online Course

 CLUSTERING
ANOMALY DETECTION

K-MEANS MULTI-CLASS CLASSIFICATION

One Class SVM > 100 Features
Fast Training, Linear Model Multi-Class Logistic Regression

PCA Based Anomaly Detection Fast Training Accuracy, Long Training Times Multi-Class Neural Network

Accuracy, Fast Training Multi-Class Decision Forest

REGRESSION Accuracy, Small Memory Footprint Multi-Class Decision Jungle

Data in Rank Order

Start Depends on Two-Class One-V-All Multiclass
Ordinal Regression
categories

Poisson Regression Predicting Event Counts

Predicting a
Fast Forest Quantile Regression Distribution TWO-CLASS CLASSIFICATION
Fast Training, Linear Two-Class Decision
Linear Regression Accuracy, Fast
Model >100 Features, Forest
Two Class SVM Training
Linear Model
Linear Model, Small Accuracy, Fast Two-Class Boosted
Bayesian Linear Regression
datasets Two-Class Averaged Fast Training, Training, LargeM Decision Tree
Perceptron Linear Model
Accuracy, Long Training Accuracy, SmallM Two Class Decision
Neural Network Regression Fast Training,
Time Two Class Logistic Jungle
Regression Linear model
Decision Forest Regression Accuracy, Fast Training >100 Features Two Class Locally Deep
Two Class Bayes Fast Training, SVM
Point Machine Linear Model
Accuracy, Fast Training, Accuracy, Long Two Class Neural
Boosted Decision Tree Regression Training Times
large Memory Network
© Jitesh Khurkhuriya – Azure ML Online Course
Thank You and Have a Great Time !

© Jitesh Khurkhuriya – Azure ML Online Course